B1245
Title: Statistical power to Bayesian assurance in superiority clinical trials
Authors: Din Chen - University of Pretoria (South Africa) [presenting]
Abstract: A well-designed clinical trial requires an appropriate sample size with adequate statistical power to address trial objectives. Statistical power is traditionally defined as the probability of rejecting the null hypothesis with a pre-specified true clinical treatment effect. This power is a conditional probability conditioned on the true but unknown effect. In practice, however, this true effect is never a fixed value but a random variable within a range of values. A paradigm shift is then to incorporate the distribution of this treatment effect from the conventional statistical power to a Bayesian statistical assurance, defined as the unconditional probability of rejecting the null hypothesis. The transition from conventional statistical power to the newly developed assurance is outlined, and the computations of assurance using the Monte-Carlo simulation-based approach in both superiority and non-inferiority clinical trials are discussed.